Intel announce self-learning Neuromorphic chip “1 million times faster” than previous designs

Intel has announced a new first-of-its-kind neural network chip which offers accelerated learning up to 1 million times faster than current designs.

The Intel Loihi chip is the result of six years of work on the specialized architecture which uses asynchronous spiking to mimic the way real synapses in the brain work. Like the chip, the brain’s neural networks relay information with pulses or spikes, modulate the synaptic strengths or weight of the interconnections based on the timing of these spikes and store these changes locally at the interconnections. Intelligent behaviours emerge from the cooperative and competitive interactions between multiple regions within the brain’s neural networks and its environment.

Intel calls the chip “self-learning”, noting it was able to learn based on various modes of feedback from the environment and did not need to be trained in the traditional way. This could help computers self-organize and make decisions based on patterns and associations.

The Intel Loihi test chip offers highly flexible on-chip learning and combines training and inference on a single chip. This allows machines to be autonomous and to adapt in real time instead of waiting for the next update from the cloud.

Researchers have demonstrated learning at a rate that is a 1 million times improvement compared with other typical spiking neural nets as measured by total operations to achieve a given accuracy when solving MNIST digit recognition problems.

Compared to technologies such as convolutional neural networks and deep learning neural networks, the Intel Loihi test chip uses many fewer resources on the same task.

The current chip emulates a total of 130,000 neurons and 130 million synapses and is up to 1,000 times more energy-efficient than general purpose computing required for typical training systems.

The chip features:

• Fully asynchronous neuromorphic many core mesh that supports a wide range of sparse, hierarchical and recurrent neural network topologies with each neuron capable of communicating with thousands of other neurons.
• Each neuromorphic core includes a learning engine that can be programmed to adapt network parameters during operation, supporting supervised, unsupervised, reinforcement and other learning paradigms.
• Fabrication on Intel’s 14 nm process technology.
• A total of 130,000 neurons and 130 million synapses.
• Development and testing of several algorithms with high algorithmic efficiency for problems including path planning, constraint satisfaction, sparse coding, dictionary learning, and dynamic pattern learning and adaptation.

The Intel Loihi test chip will be shared with leading university and research institutions in the first half of 2018 with a focus on advancing AI. Intel has not revealed plans to commercialize the chip yet.